Detail publikace

The Q-learned controller for small asynchronous electromotor

Originální název

The Q-learned controller for small asynchronous electromotor

Anglický název

The Q-learned controller for small asynchronous electromotor

Jazyk

en

Originální abstrakt

The contribution is focused on the application of two staged Q-learning on asynchronous electromotor (AE) control. Tutorage stage uses conventional Q-learning procedure applied on Q function created during the prelearning stage. Tutorage is expected to be processed on real AE plant and its use for controller adaptation on possible system parameters changes is not excluded. Based on the tests the best grids of Q-function table for 2-D state and 3-D state were selected and obtained policies performance was compared with the performance of AE controlled by PID, which was tuned through Ziegler-Nichols methodology.

Anglický abstrakt

The contribution is focused on the application of two staged Q-learning on asynchronous electromotor (AE) control. Tutorage stage uses conventional Q-learning procedure applied on Q function created during the prelearning stage. Tutorage is expected to be processed on real AE plant and its use for controller adaptation on possible system parameters changes is not excluded. Based on the tests the best grids of Q-function table for 2-D state and 3-D state were selected and obtained policies performance was compared with the performance of AE controlled by PID, which was tuned through Ziegler-Nichols methodology.

BibTex


@inproceedings{BUT16176,
  author="Tomáš {Marada} and Tomáš {Březina} and Vladislav {Singule}",
  title="The Q-learned controller for small asynchronous electromotor",
  annote="The contribution is focused on the application of two staged Q-learning on asynchronous electromotor (AE) control. Tutorage stage uses conventional Q-learning procedure applied on Q function created during the prelearning stage. Tutorage is expected to be processed on real AE plant and its use for controller adaptation on possible system parameters changes is not excluded. Based on the tests the best grids of Q-function table for 2-D state and 3-D state were selected and obtained policies performance was compared with the performance of AE controlled by PID, which was tuned through 
Ziegler-Nichols methodology.",
  address="Fakulta Mechatroniky TnU AD v spolupráci s firmou GC TECH Ing. Peter Gerši",
  booktitle="Mechatronika 2005",
  chapter="16176",
  institution="Fakulta Mechatroniky TnU AD v spolupráci s firmou GC TECH Ing. Peter Gerši",
  journal="Nezařazené články",
  year="2005",
  month="may",
  pages="167",
  publisher="Fakulta Mechatroniky TnU AD v spolupráci s firmou GC TECH Ing. Peter Gerši",
  type="conference paper"
}